Appendix

Unequal Representation: Evidence from a Census of Legislator Contacts with US Federal Agencies

Theory

Synthetic figures, ignoring baseline levels of each type.

2x2

library(tidyverse)
x2 <- read_csv(here::here("data", "2x2.csv"))
# 
# x2 %>% 
#   ggplot() +
#   aes(x = x, y = new, label = lab) +
#   geom_line(aes(y = baseline), linetype = 2, alpha = .5) + 
#   geom_line() +
#   geom_ribbon(aes(ymax = new, ymin = baseline), alpha = .2) + 
#   facet_wrap("outcome", nrow = 2) + 
#   geom_label() +  
#   geom_text(aes(label = lab2, y = baseline), check_overlap = T, hjust = 0, vjust = -.2) +
#   labs(y = "Change in Level Provided\n(e.g., the number of requests to agencies)",
#        title = "") + 
#   theme_void() + 
#   xlim(-.2,1.2) + 
#   ylim(-1.1,3.1) + 
#   theme(panel.border = element_rect(fill=NA),
#         axis.title.y = element_text(angle = 90))



x2 %>% 
  ggplot() +
  aes(x = x, y = new, label = lab) +
  geom_line(aes(y = baseline), linetype = 2, alpha = .5) + 
  geom_line() +
  geom_ribbon(aes(ymax = new, ymin = baseline), alpha = .2) + 
  facet_grid(priority ~ capacity) +
  geom_label() +  
  geom_text(y = 0, 
            label = "Baseline levels for new legislator\n(may differ between service and policy work)",
            aes(x = ifelse(capacity == "Capacity DOES affect behavior" & priority == "Shifting priorities DO affect behavior", 0.5, NA))) +
  labs(y = "Change in Level Provided\n(e.g., the number of requests to agencies)",
       title = "",
       x = "") + 
  #theme_minimal() + 
  xlim(-.2,1.2) + 
  ylim(-1.1,3.1) + 
  theme(panel.grid = element_blank(),
        panel.border = element_rect(fill=NA),
        axis.text = element_blank(),
        axis.ticks = element_blank(),
        panel.background = element_blank())

Divergent Predictions for the Change in Levels of Constituency Service and Policy Work as Legislators Gain Experience and Power

# x2 %>% 
#   ggplot() +
#   aes(x = x, y = new, label = lab) +
#   #geom_point(aes(y = baseline), alpha = .5) + 
#   #geom_line() +
#   geom_segment(aes(yend = baseline, xend = x), alpha = .2, size = 20) + 
#   facet_wrap("outcome", nrow = 2) + 
#   geom_label() +  
#   geom_text(aes(label = lab2, y = baseline), check_overlap = T, hjust = 0, vjust = 1.5) +
#   labs(y = "Change in Level Provided\n(e.g., the number of requests to agencies)",
#        title = "") + 
#   theme_void() + 
#   xlim(-.1,1.1) + 
#   ylim(-1.1,3.1) + 
#   theme(panel.border = element_rect(fill=NA),
#         axis.title.y = element_text(angle = 90))

1x2

# just 
x2 %>% 
  filter(outcome %in% c("3. Only Increased Capacity Affects Behavior", "2. Only Shifting Priorities Affect Behavior")) %>% 
  mutate(outcome = str_remove(outcome, "[1-9].*Only ")) %>%
  ggplot() +
  aes(x = x, y = new, label = lab) +
  geom_line(aes(y = baseline), linetype = 2, alpha = .5) + 
  geom_line() +
  geom_ribbon(aes(ymax = new, ymin = baseline), alpha = .2) + 
  facet_wrap("outcome", nrow = 1) + 
  geom_label() +  
  geom_text(y = 0, 
            vjust = 1.1,
            label = "Baseline levels for a new legislator\n(may differ between service & policy work)",
            aes(x = ifelse(outcome == "Increased Capacity Affects Behavior", 0.5, NA))) +
  labs(y = "Change in Level Provided\n(e.g., the number of\nrequests to agencies)",
       title = "") + 
  theme_void() + 
  xlim(-.3,1.3) + 
  ylim(-1.1,1.1) + 
  theme(panel.border = element_rect(fill=NA),
        axis.title.y = element_text(angle = 90))

Divergent Predictions for the Change in Levels of Constituency Service and Policy Work as Legislators Gain Experience and Power

Data

FOIA <- read_csv(here("data/_FOIA_response_table.csv")) %>% select(-ends_with("1")) %>% 
  mutate(Department = Department %>% str_remove("Department of ")) 

FOIA %>% 
  kablebox_long()

Department	Components FOIAed	Records received	Coded	Observations
Agriculture	29	29	11	9516
Commerce	19	18	10	8038
Defense	49	13	7	9739
Education	1	1	1	4689
Energy	8	2	1	6580
Health and Human Services	15	10	7	104145
Homeland Security	14	13	12	39633
Housing and Urban Development	2	1	1	33968
Justice	23	5	3	2611
Labor	22	12	8	53341
State	1	0	0	0
the Interior	11	8	6	6079
the Treasury	7	5	5	23869
Transportation	10	7	6	26787
Veterans Affairs	6	3	2	77842
Independent Agencies	77	47	30	81053
Total	294	174	110	487890

FOIA %>% select(-Coded) %>%
  rename(N = Observations) %>% 
  kable(format = "latex", booktabs = T) %>%
  row_spec(row = 16, hline_after = TRUE) %>%
  row_spec(row = 17, bold = T) %>% 
  #kable_styling(font_size = 9, full_width = TRUE) %>% 
       write_file(here::here("tables", "FOIA.tex")) 

# can't cache data

# raw data 
load(here::here("data", "all_contacts2.Rdata"))
all_contacts2 <- all_contacts #FIXME 

load(here::here("data", "all_contacts_paper.Rdata"))

all_contacts2 %<>% mutate(type = ifelse(TYPE %in% c(1,2,3), "Constituency\nService", NA) ) %>%
  mutate(type = ifelse(TYPE %in% c(4,5), "Policy\nWork", type) )

all_contacts2$Type %<>% str_replace("Indiv. Constituent" , "Constituent (Individual)") %>% 
  str_replace("Corp. Constituent" , "Constituent (Corporate)")%>% 
  str_replace("501c3 or Local Gov." , "Constituent (501c3 or Local Gov.)")%>% 
  str_replace("Corp. Policy" , "Policy (Corporate)")%>% 
  str_replace("^Policy$" , "Policy (General)")


load(here::here("data", "dcounts_min.Rdata"))
load(here::here("data", "dcounts_tenure.Rdata"))
load(here::here("data", "dcounts_ratio.Rdata"))
load(here::here("data", "dcounts_per_district.Rdata"))


load(here::here("data", "members.Rdata"))

members %<>% mutate(member_state = member_state %>%
                     str_replace("\\) ", "-") %>%
                     str_c(")")) 


congress_years2<- function(congress){
        years<- c(congress*2 + 1787, congress*2 + 1788 ) %>% str_c(collapse = ";", sep = ";")
        return(years)
}

# FIXME members data should have years so we don't merge in observations for members in the january (final days of their last term)
# members2 <- members %>%
# mutate(year = congress %>%  congress_years2() %>% str_c(collapse = ";"))

# CAUTION THIS IS WRONG AND WILL PRINT OUT ALL YEARS IN DATA ^2
# members2$year[1]

load(here::here('data', 'agency_vars.Rdata'))

congress_years<- function(congress){
        years<- c(congress*2 + 1787, congress*2 + 1788 )
        return(years)
}

year_congress<- function(year){
    return(floor((year - 1787)/2))
}



d <- dcounts_min %>% 
  ungroup() %>% 
  mutate(type = ifelse(TYPE %in% c(1,2,3), "Constituency\nService", NA) ) %>%
  mutate(type = ifelse(TYPE %in% c(4,5), "Policy\nWork", type) )

type <- d %>% 
  group_by(type) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup()


type %>% 
  rename(type2 = type) %>%
  mutate(percent = n/sum(type$n[1:2])*100) %>% kablebox()

type2	n	percent
Constituency Service	299741	80.73723
Policy Work	71514	19.26277
NA	99832	26.89041

type5 <- d %>% 
  group_by(TYPE) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() 

type5 %<>% 
  mutate(percent = n/sum(type5$n[2:6])*100)

type5 %>% kablebox()

TYPE	n	percent
0	4787	1.2894102
1	240639	64.8177129
2	23794	6.4090719
3	35308	9.5104443
4	13112	3.5318043
5	58402	15.7309666
6	2416	0.6507656
NA	92629	24.9502364

type2017 <- d %>% 
  filter(year> 2006, year < 2018) %>%
  group_by(type) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>%
  ungroup()

coded <- d %>% 
  mutate(coded = !is.na(type)) %>% 
  group_by(coded) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) 

coded_ratio = coded$n[2]/(sum(coded$n))

multiplier = sum(coded$n)/coded$n[2]

percent <- function(x){
  y <- x*100 
  y %<>% round() 
  return(y)}
# percent(.434534523453)

d %<>% filter(chamber != "President")

d %<>% mutate(congress = year_congress(year))




dcounts <- d %>% left_join(members)


type2xchair <- dcounts %>% 
  group_by(type, chair) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  drop_na(type, chair) %>% 
  group_by(chair) %>% 
  mutate(percent = n / sum(n) *100) 

type2xchair%>% 
  kablebox()

type	chair	n	percent
Constituency Service	0	264161	81.70214
Constituency Service	1	35230	74.22780
Policy Work	0	59161	18.29786
Policy Work	1	12232	25.77220

type2xprestigechair <- dcounts %>% 
  group_by(type, prestige_chair) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  drop_na() %>% 
  group_by(prestige_chair) %>% 
  mutate(percent = n / sum(n) *100) 

type2xprestigechair%>% 
  kablebox()

type	prestige_chair	n	percent
Constituency Service	0	286381	80.85725
Constituency Service	1	13010	78.35933
Policy Work	0	67800	19.14275
Policy Work	1	3593	21.64067

# prestige committee
type2xprestige <- dcounts %>% 
  group_by(type, prestige) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  drop_na() %>% 
  group_by(prestige) %>% 
  mutate(percent = n / sum(n) *100) 


type2xprestigechair%>% 
  kablebox()

type	prestige_chair	n	percent
Constituency Service	0	286381	80.85725
Constituency Service	1	13010	78.35933
Policy Work	0	67800	19.14275
Policy Work	1	3593	21.64067

type2xchamber <- dcounts %>%
  group_by(type, chamber) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  drop_na() %>% 
  group_by(chamber) %>% 
  mutate(percent = n / sum(n) *100) 


# year
type2xyear <- dcounts %>% mutate(years = year  - yearelected) %>% 
  group_by(type, years) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  drop_na() %>% 
  group_by(years) %>% 
  mutate(percent = n / sum(n) *100) 

type2xyear %>% 
  kablebox()

type	years	n	percent
Constituency Service	1	145812	79.88954
Constituency Service	2	153567	81.57480
Policy Work	1	36705	20.11046
Policy Work	2	34686	18.42520

# chamber averages 
dcounts %>%
  group_by(chamber, year, bioname) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  arrange(-n) %>%
  group_by(chamber) %>% 
  summarise(mean_per_year = mean(n) %>% round(0)) %>% 
  kablebox()

chamber	mean_per_year
House	52
Senate	156

# per member 
dcounts %>% 
  group_by(bioname, year) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  arrange(-n) %>%
  group_by(bioname) %>% 
  mutate(mean_per_year = mean(n) %>% round(0)) %>% 
  kablebox_long()

bioname	year	n	mean_per_year
WARNER, Mark	2014	1032	453
WEBB, James H. (Jim)	2010	849	517
CHAMBLISS, Saxby	2009	800	294
McCAIN, John Sidney, III	2015	784	380
SCHUMER, Charles Ellis (Chuck)	2012	732	486
BYRD, Robert Carlyle	2009	691	510
NELSON, Clarence William (Bill)	2010	668	455
MIKULSKI, Barbara Ann	2010	657	393
SCHUMER, Charles Ellis (Chuck)	2010	650	486
GRASSLEY, Charles Ernest	2010	649	383
WARNER, Mark	2015	640	453
BYRD, Robert Carlyle	2008	621	510
McCAIN, John Sidney, III	2016	620	380
GILLIBRAND, Kirsten	2013	611	332
WEBB, James H. (Jim)	2009	603	517
WEBB, James H. (Jim)	2008	599	517
NELSON, Clarence William (Bill)	2009	596	455
GRASSLEY, Charles Ernest	2011	595	383
NELSON, Clarence William (Bill)	2011	580	455
GILLIBRAND, Kirsten	2012	572	332
DURBIN, Richard Joseph	2009	561	272
SCHUMER, Charles Ellis (Chuck)	2009	561	486
LUGAR, Richard Green	2010	558	410
SCHUMER, Charles Ellis (Chuck)	2016	554	486
NELSON, Clarence William (Bill)	2013	553	455
LUGAR, Richard Green	2011	541	410
MIKULSKI, Barbara Ann	2013	541	393
MANCHIN, Joe, III	2011	525	260
NELSON, Clarence William (Bill)	2012	522	455
MIKULSKI, Barbara Ann	2011	521	393
SCHUMER, Charles Ellis (Chuck)	2011	507	486
WOLF, Frank Rudolph	2010	502	377
GILLIBRAND, Kirsten	2011	501	332
WOLF, Frank Rudolph	2009	496	377
GILLIBRAND, Kirsten	2014	492	332
WARNER, Mark	2013	487	453
WARNER, Mark	2016	486	453
SCHUMER, Charles Ellis (Chuck)	2013	482	486
WARNER, Mark	2011	482	453
WOLF, Frank Rudolph	2011	480	377
MIKULSKI, Barbara Ann	2014	478	393
SHELBY, Richard C.	2011	476	284
RUBIO, Marco	2017	475	264
LUGAR, Richard Green	2009	466	410
KIRK, Mark Steven	2011	465	220
McCAIN, John Sidney, III	2010	465	380
SCHUMER, Charles Ellis (Chuck)	2014	464	486
GRASSLEY, Charles Ernest	2012	462	383
DURBIN, Richard Joseph	2008	458	272
WARNER, Mark	2012	456	453
NELSON, Clarence William (Bill)	2016	455	455
TOOMEY, Patrick Joseph	2015	453	343
SPECTER, Arlen	2008	451	332
NELSON, Clarence William (Bill)	2014	450	455
OBAMA, Barack	2008	450	308
WEBB, James H. (Jim)	2011	449	517
KIRK, Mark Steven	2013	442	220
BROWN, Sherrod	2012	440	284
TOOMEY, Patrick Joseph	2016	433	343
SPECTER, Arlen	2010	430	332
CHAMBLISS, Saxby	2008	429	294
LUGAR, Richard Green	2008	429	410
GRASSLEY, Charles Ernest	2009	425	383
BROWN, Sherrod	2011	423	284
GILLIBRAND, Kirsten	2016	422	332
McCAIN, John Sidney, III	2017	422	380
CASEY, Robert (Bob), Jr.	2012	419	260
McCAIN, John Sidney, III	2009	417	380
MIKULSKI, Barbara Ann	2012	417	393
SCHUMER, Charles Ellis (Chuck)	2008	412	486
GRASSLEY, Charles Ernest	2015	410	383
MIKULSKI, Barbara Ann	2009	410	393
BOOKER, Cory Anthony	2015	408	265
RUBIO, Marco	2016	408	264
SHELBY, Richard C.	2009	403	284
CORNYN, John	2009	402	265
GRASSLEY, Charles Ernest	2013	401	383
SCHUMER, Charles Ellis (Chuck)	2015	399	486
WOLF, Frank Rudolph	2012	397	377
SHELBY, Richard C.	2012	396	284
REID, Harry	2011	391	212
DURBIN, Richard Joseph	2010	387	272
SCHUMER, Charles Ellis (Chuck)	2017	387	486
SHELBY, Richard C.	2008	386	284
BURR, Richard M.	2016	383	205
WOLF, Frank Rudolph	2008	383	377
NELSON, Clarence William (Bill)	2007	382	455
GILLIBRAND, Kirsten	2010	380	332
CASEY, Robert (Bob), Jr.	2011	377	260
MANCHIN, Joe, III	2014	377	260
HAGAN, Kay	2014	375	274
GILLIBRAND, Kirsten	2015	374	332
MANCHIN, Joe, III	2012	374	260
GRAHAM, Lindsey O.	2014	373	202
TOOMEY, Patrick Joseph	2017	373	343
BYRD, Robert Carlyle	2010	372	510
GRASSLEY, Charles Ernest	2014	372	383
NELSON, Clarence William (Bill)	2008	372	455
CASEY, Robert (Bob), Jr.	2013	368	260
GRASSLEY, Charles Ernest	2016	368	383

# per member HOUSE
dcounts %>% 
  filter(chamber == "House") %>% 
  group_by(bioname, year) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  arrange(-n) %>%
  group_by(bioname) %>% 
  mutate(mean_per_year = mean(n) %>% round(0)) %>% 
  kablebox_long()

bioname	year	n	mean_per_year
WOLF, Frank Rudolph	2010	502	377
WOLF, Frank Rudolph	2009	496	377
WOLF, Frank Rudolph	2011	480	377
WOLF, Frank Rudolph	2012	397	377
WOLF, Frank Rudolph	2008	383	377
FRANK, Barney	2010	346	290
GARRETT, Scott	2014	342	145
FRANK, Barney	2009	335	290
McCAUL, Michael T.	2012	334	202
GARRETT, Scott	2016	333	145
COMSTOCK, Barbara J.	2016	332	208
McCAUL, Michael T.	2010	329	202
SHIMKUS, John M.	2010	326	156
FRANK, Barney	2011	323	290
SESTAK, Joe	2010	313	216
McCAUL, Michael T.	2009	308	202
RODRIGUEZ, Ciro D.	2008	306	215
DENT, Charles W.	2010	303	132
SHIMKUS, John M.	2009	303	156
GOODLATTE, Robert William	2014	302	172
EDWARDS, Donna F.	2009	295	96
CARTWRIGHT, Matt	2014	287	139
CARTER, Buddy	2016	284	191
RODRIGUEZ, Ciro D.	2009	284	215
McCAUL, Michael T.	2008	283	202
SHIMKUS, John M.	2008	280	156
WOLF, Frank Rudolph	2013	279	377
McCAUL, Michael T.	2011	273	202
PAUL, Ronald Ernest	2010	268	156
FILNER, Bob	2010	264	131
VAN HOLLEN, Christopher	2011	262	157
FRANK, Barney	2012	259	290
GOODLATTE, Robert William	2013	257	172
WOLF, Frank Rudolph	2014	256	377
ALEXANDER, Rodney	2010	254	204
ALEXANDER, Rodney	2012	253	204
FRANK, Barney	2008	253	290
PAUL, Ronald Ernest	2009	252	156
DENT, Charles W.	2011	248	132
SESTAK, Joe	2009	247	216
ALEXANDER, Rodney	2014	245	204
COMSTOCK, Barbara J.	2015	245	208
FORBES, J. Randy	2015	237	154
VAN HOLLEN, Christopher	2010	236	157
GOODLATTE, Robert William	2016	235	172
RODRIGUEZ, Ciro D.	2010	234	215
CHAFFETZ, Jason	2016	232	62
ALEXANDER, Rodney	2013	228	204
STEFANIK, Elise M	2016	227	181
STEFANIK, Elise M	2017	224	181
WOLF, Frank Rudolph	2007	223	377
FRANK, Barney	2007	222	290
TIPTON, Scott	2015	222	111
GARRETT, Scott	2015	217	145
SHEA-PORTER, Carol	2014	217	76
MURPHY, Timothy	2014	216	87
CARTER, Buddy	2015	215	191
RAHALL, Nick Joe, II	2013	214	126
MILLER, Jefferson B. (Jeff)	2014	212	65
SCOTT, Austin	2016	212	120
ALEXANDER, Rodney	2011	211	204
FORBES, J. Randy	2016	210	154
CARTWRIGHT, Matt	2015	209	139
DAVIS, Rodney	2018	209	136
VISCLOSKY, Peter	2012	209	145
ACKERMAN, Gary Leonard	2010	207	116
MURPHY, Patrick	2014	207	78
GRIFFITH, H. Morgan	2015	206	143
KINGSTON, Jack	2012	206	142
BENISHEK, Dan	2013	205	108
DAVIS, Rodney	2016	205	136
MOORE, Dennis	2009	205	111
PALLONE, Frank, Jr.	2016	205	110
GOODLATTE, Robert William	2012	204	172
VISCLOSKY, Peter	2009	204	145
ALEXANDER, Rodney	2009	203	204
STEFANIK, Elise M	2015	200	181
GRAHAM, Gwendolyn (Gwen)	2016	198	190
PALLONE, Frank, Jr.	2017	198	110
SCOTT, Austin	2014	198	120
VAN HOLLEN, Christopher	2009	197	157
COLLINS, Chris	2018	196	110
LAMPSON, Nicholas V.	2008	196	124
DAVIS, Lincoln	2009	195	156
DAVIS, Rodney	2017	195	136
OWENS, William	2013	192	138
RICE, Kathleen Maura	2015	192	112
WITTMAN, Robert J.	2015	192	104
GRIFFITH, H. Morgan	2014	190	143
DAVIS, Lincoln	2008	189	156
GOODLATTE, Robert William	2015	189	172
GARRETT, Scott	2013	188	145
OWENS, William	2011	188	138
SARBANES, John	2010	188	93
SCOTT, Robert C.	2015	188	101
UPTON, Frederick Stephen	2012	188	96
UPTON, Frederick Stephen	2013	188	96
FORBES, J. Randy	2013	187	154
VISCLOSKY, Peter	2013	187	145
PITTS, Joseph R.	2009	185	109

# per member VA-10
dcounts %>% 
  filter(cqlabel == "(VA-10)") %>% 
  group_by(bioname, year) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  arrange(year) %>%
  group_by(bioname) %>% 
  mutate(mean_per_year = mean(n) %>% round(0)) %>% 
  kablebox_long()

bioname	year	n	mean_per_year
WOLF, Frank Rudolph	2007	223	377
WOLF, Frank Rudolph	2008	383	377
WOLF, Frank Rudolph	2009	496	377
WOLF, Frank Rudolph	2010	502	377
WOLF, Frank Rudolph	2011	480	377
WOLF, Frank Rudolph	2012	397	377
WOLF, Frank Rudolph	2013	279	377
WOLF, Frank Rudolph	2014	256	377
COMSTOCK, Barbara J.	2015	245	208
COMSTOCK, Barbara J.	2016	332	208
COMSTOCK, Barbara J.	2017	160	208
COMSTOCK, Barbara J.	2018	97	208

dcounts %>% 
  group_by(bioname, year) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  arrange(n) %>%
  group_by(bioname) %>% 
  mutate(mean_per_year = mean(n) %>% round(0)) %>% 
  #FIXME 
  filter(n > 4) %>% 
  kablebox_long()

bioname	year	n	mean_per_year
AKIN, W. Todd	2008	5	16
AMODEI, Mark E.	2018	5	14
BARROW, John	2007	5	65
BIDEN, Joseph Robinette, Jr.	2009	5	27
BISHOP, Robert (Rob)	2007	5	29
CROWLEY, Joseph	2018	5	33
DAVIS, Danny K.	2018	5	25
GARRETT, Thomas Alexander Jr.	2018	5	13
HUIZENGA, Bill	2018	5	26
JINDAL, Bobby	2008	5	22
KENNEDY, Edward Moore (Ted)	2010	5	128
KHANNA, Rohit	2018	5	10
LAMB, Conor	2017	5	6
LEVIN, Sander Martin	2018	5	40
LEWIS, John R.	2018	5	54
LYNCH, Stephen F.	2018	5	26
MARSHALL, Roger Wayne	2018	5	10
MULVANEY, Mick	2017	5	20
NUNNELEE, Alan	2015	5	34
POMEROY, Earl Ralph, III	2007	5	23
SALAZAR, Kenneth Lee	2010	5	42
SÁNCHEZ, Linda T.	2018	5	21
SHULER, Heath	2007	5	27
TITUS, Alice (Dina)	2018	5	30
TORRES, Norma Judith	2018	5	14
VARGAS, Juan	2018	5	32
WATERS, Maxine	2018	5	23
WELLER, Gerald C. (Jerry)	2007	5	8
ABERCROMBIE, Neil	2010	6	18
BACHMANN, Michele	2007	6	31
BIDEN, Joseph Robinette, Jr.	2010	6	27
BUTTERFIELD, George Kenneth, Jr. (G.K.)	2018	6	28
CALVERT, Ken	2018	6	38
CLYBURN, James Enos	2008	6	16
GRAVES, Samuel	2018	6	56
GREEN, Al	2017	6	28
KILPATRICK, Carolyn Cheeks	2007	6	13
KUSTOFF, David	2018	6	12
LaHOOD, Darin	2015	6	31
LAWRENCE, Brenda L	2018	6	29
LEWIS, Jason Mark	2018	6	8
LOWENTHAL, Alan	2018	6	46
MARCHANT, Kenny	2007	6	23
MARCHANT, Kenny	2018	6	23
MARSHALL, Jim	2007	6	48
MEEHAN, Martin Thomas	2007	6	3
MOONEY, Alex X.	2018	6	13
NAPOLITANO, Grace Flores	2007	6	15
NAPOLITANO, Grace Flores	2018	6	15
PALMER, Gary James	2018	6	18
PETERSON, Collin Clark	2018	6	26
PRICE, David Eugene	2018	6	46
ROYBAL-ALLARD, Lucille	2018	6	27
RUSH, Bobby L.	2018	6	43
SALAZAR, Kenneth Lee	2009	6	42
SESSIONS, Jefferson Beauregard III (Jeff)	2017	6	120
SMITH, Adrian	2018	6	24
STEWART, Chris	2018	6	18
WICKER, Roger F.	2007	6	55
ADAMS, Alma	2018	7	25
BECERRA, Xavier	2008	7	18
BISHOP, Sanford Dixon, Jr.	2007	7	38
BONAMICI, Suzanne	2018	7	22
CHENEY, Liz	2018	7	11
ESTES, Ron	2017	7	14
GABBARD, Tulsi	2018	7	42
GARAMENDI, John	2009	7	60
GRIJALVA, Raúl M.	2018	7	63
GUTIÉRREZ, Luis V.	2007	7	16
HARMAN, Jane L.	2011	7	20
ISSA, Darrell	2018	7	81
JEFFRIES, Hakeem	2018	7	14
JONES, Stephanie Tubbs	2007	7	22
KNIGHT, Steve	2018	7	28
LUCAS, Frank D.	2018	7	16
McCOLLUM, Betty	2008	7	30
PORTER, Jon C.	2007	7	8
PRYCE, Deborah D.	2008	7	9
ROONEY, Thomas J.	2018	7	32
SEWELL, Terri	2018	7	45
SULLIVAN, John	2008	7	23
SWALWELL, Eric	2018	7	36
TURNER, Robert L.	2011	7	26
WEXLER, Robert	2010	7	64
AKIN, W. Todd	2007	8	16
BARRASSO, John A.	2007	8	52
BEATTY, Joyce	2018	8	33
BECERRA, Xavier	2010	8	18
BISHOP, Robert (Rob)	2008	8	29
BLUMENAUER, Earl	2018	8	34
BOEHNER, John Andrew	2016	8	56
BUCHANAN, Vernon G.	2007	8	43
CLARKE, Yvette Diane	2007	8	31
CLAY, William Lacy, Jr.	2018	8	56
CONAWAY, K. Michael	2018	8	23
COOPER, James Hayes Shofner	2018	8	19
CORREA, Jose Luis	2018	8	13
COSTA, Jim	2018	8	28
CURTIS, John R.	2018	8	4
DINGELL, Debbie	2018	8	26

Example

Rep. Barbara Comstock worked in Rep. Frank Wolf’s congressional office for five years, described him as a “longtime mentor,” and claimed that Wolf first urged her to run for office. He endoresed and campaigned with her.

Notice several patterns - dip in constituncy service when Wolf is no longer running for re-eleiction - further dip with transition - quick rebound as Comstock gets her office in order - inheriting a highly-effective office, Comstock is far above average house member

cq <- "(VA-10)"

dcounts %>% 
  group_by(member_state, year, chamber, cqlabel) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  group_by(chamber) %>% 
  mutate(mean_per_chamber = mean(n) %>% round(0)) %>% 
  filter(cqlabel == cq, year > 2007, year <2017) %>% 
  ggplot() +
  aes(x = year, y = n, fill = member_state) + 
  geom_col(alpha = .5, position = "dodge") +
  geom_hline(aes(yintercept = mean_per_chamber), linetype = 2) + 
  geom_text(aes(y = mean_per_chamber), x = 2007.5, label = "House mean", min.segment.length = 0, size = 3, fill = NA, vjust = -.3, hjust = 0) + 
  scale_x_continuous(breaks = seq(2008, 2016, 1)) +
  labs(x = "",
       y = "Number of Requests", 
       fill = cq %>% str_remove_all("\\(|\\)")) + 
  theme_minimal() + 
  theme(panel.grid.major.x =  element_blank(),
        panel.grid.minor.x =  element_blank(),
        axis.text.x = element_text(angle = 45))

cq <- "(NY-07)"
CHAMBER <- "House"
dcounts %>% 
  group_by(member_state, year, chamber, cqlabel) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  group_by(chamber) %>% 
  mutate(mean_per_chamber = mean(n) %>% round(0)) %>% 
  ggplot() +
  aes(x = year, y = n, 
      fill = member_state %>% str_replace("-.*", "\\)")) + 
  geom_col(alpha = .5, position = "dodge") +
  geom_hline(aes(yintercept = mean_per_chamber), linetype = 2) + 
  geom_text(aes(y = mean_per_chamber), x = 2007.5, label = "Chamber average", min.segment.length = 0, size = 3, fill = NA, vjust = 1.3, hjust = 0) + 
  scale_x_continuous(breaks = seq(2007, 2017, 1)) +
  labs(x = "",
       y = "Number of Requests", 
       fill = str_c(CHAMBER, "-", cq) %>% str_remove_all("\\(|\\)") ) + 
  theme_minimal() + 
  theme(panel.grid.major.x =  element_blank(),
        panel.grid.minor.x =  element_blank(),
        axis.text.x = element_text(angle = 45))

cq <- "(NY-14)"
CHAMBER <- "House"
dcounts %>% 
  group_by(member_state, year, chamber, cqlabel) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  group_by(chamber) %>% 
  mutate(mean_per_chamber = mean(n) %>% round(0)) %>% 
  filter(cqlabel == cq, year > 2007, year <2019) %>% 
  ggplot() +
  aes(x = year, y = n, 
      fill = member_state %>% str_replace("-.*", "\\)")) + 
  geom_col(alpha = .5, position = "dodge") +
  geom_hline(aes(yintercept = mean_per_chamber), linetype = 2) + 
  geom_text(aes(y = mean_per_chamber), x = 2007.5, label = "Chamber average", min.segment.length = 0, size = 3, fill = NA, vjust = 1.3, hjust = 0) + 
  scale_x_continuous(breaks = seq(2008, 2017, 1)) +
  labs(x = "",
       y = "Number of Requests", 
       fill = str_c(CHAMBER, "-", cq) %>% str_remove_all("\\(|\\)") ) + 
  theme_minimal() + 
  theme(panel.grid.major.x =  element_blank(),
        panel.grid.minor.x =  element_blank(),
        axis.text.x = element_text(angle = 45))

case_study <- function(cq){

  CHAMBER <- filter(members, cqlabel == cq) %>% 
    .$chamber %>% .[1]

dcounts %>% 
  group_by(member_state, year, chamber, cqlabel) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
  ungroup() %>% 
  group_by(chamber) %>% 
  mutate(mean_per_chamber = mean(n) %>% round(0)) %>% 
  filter(cqlabel == cq, year > 2007, year <2017) %>% 
  ggplot() +
  aes(x = year, y = n, 
      fill = member_state %>% str_replace("-.*", "\\)")) + 
  geom_col(alpha = .5, position = "dodge") +
  geom_hline(aes(yintercept = mean_per_chamber), linetype = 2) + 
  geom_text(aes(y = mean_per_chamber), x = 2007.5, label = "Chamber average", min.segment.length = 0, size = 3, fill = NA, vjust = 1.3, hjust = 0) + 
  scale_x_continuous(breaks = seq(2008, 2016, 1)) +
  labs(x = "",
       y = "Number of Requests", 
       fill = str_c(CHAMBER, "-", cq) %>% str_remove_all("\\(|\\)") ) + 
  theme_minimal() + 
  theme(panel.grid.major.x =  element_blank(),
        panel.grid.minor.x =  element_blank(),
        axis.text.x = element_text(angle = 45))
  
ggsave(str_c("figs/districts/", cq, ".png"), 
       width = 5.3,
  height = 2.5,
  dpi = 1000)
} 

case_study("(NY-14)")

cases <- unique(members$cqlabel) #%>% head()

# map(cases, case_study )

---

Between February 2017 and February 2021, we received data on 471087 instances of members of Congress contacting federal agencies. We focus on requests made from 2007-2017, resulting in a data set of 452323 contacts from members of Congress with federal agencies.

Requests by percentile

dcounts %>% 
    group_by(member_state, chamber, year) %>% summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>%
  group_by(member_state, chamber) %>% 
  summarise(mean = mean(n)) %>% 
  ungroup() %>% 
    group_by(chamber) %>% 
  mutate(Percentile = ntile(mean, 100),
         rank = dense_rank(-mean)) %>% 
  arrange(-Percentile) %>% 
  dplyr::select(member_state, Percentile, mean, chamber,rank) %>% distinct() %>% 
    ggplot() + 
    geom_col(aes(x = Percentile, y = mean), 
             color = "grey", 
             width = .000001,
             fill = "black",
             position = "dodge") + 
  geom_point(aes(x = Percentile, y = mean), color = "light blue") + 
  geom_text(aes(x = Percentile, y = mean, label = ifelse(rank <14 | Percentile < 30, member_state, "")), check_overlap = T, 
                  size = 2,
                  hjust = "inward") + 
  theme_minimal() + 
  labs(#title = "Average Legislator Requests per Year by Percentile",
       x = "Percentile Within Chamber",
       y = "Average Requests per Year") +     facet_grid(. ~ chamber , scales = "free_y") 

Requests by year

dcounts %<>% 
  group_by(bioname, year) %>% 
  mutate(per_bioname_year = sum(per_icpsr_chamber_year_agency_type)) %>%
  ungroup() %>% 
  group_by(year) %>%
  #mutate(per_bioname_year = per_bioname_year - mean(per_bioname_year)) %>% 
  ungroup() 

schumer2013 <- dcounts %>% filter(bioname == "SCHUMER, Charles Ellis (Chuck)",
                   year == 2013) %>% .$per_bioname_year %>% unique()

cruz2013 <- dcounts %>% filter(bioname == "CRUZ, Rafael Edward (Ted)",
                   year == 2013) %>% .$per_bioname_year %>% unique()

mcconnell2013 <- dcounts %>% filter(bioname == "McCONNELL, Addison Mitchell (Mitch)",
                   year == 2013) %>% .$per_bioname_year %>% unique()


p <- dcounts %>% 
  filter(year > 2006, year <2018) %>%
  distinct(bioname, per_bioname_year, year, chamber) %>% 
  filter(per_bioname_year > 0) %>% 
  ggplot() + 
  geom_line(aes(x = year, 
                y = per_bioname_year, 
                group = bioname, 
                color = chamber), 
            alpha = .1) +
  geom_line(aes(x = year, 
                group = bioname, 
                color = chamber,
                y = ifelse(bioname == "SCHUMER, Charles Ellis (Chuck)", per_bioname_year, NA))) + 
  geom_label(x = 2013, 
             y = schumer2013, 
             label = "Chuck Schumer") +
  geom_line(aes(x = year, group = bioname, color = chamber,
                y = ifelse(bioname == "CRUZ, Rafael Edward (Ted)", per_bioname_year, NA))) + 
  geom_label(x = 2013, 
             y = cruz2013, 
             label = "Ted Cruz") +
  geom_line(aes(x = year, group = bioname, color = chamber,
                y = ifelse(bioname == "McCONNELL, Addison Mitchell (Mitch)", per_bioname_year, NA))) + 
  geom_label(x = 2013, 
             y = mcconnell2013, 
             label = "Mitch McConnell") +
  labs(x = "",
       color = " ",
       y = paste("Requests per Legislator per Year"),
       title = "") +
  scale_x_continuous(breaks = unique(dcounts$year)) + 
  scale_color_viridis_d(begin = 0, end = .6, option = "cividis", direction = -1) +
  theme_minimal() + 
  theme(panel.background = element_blank(),
        axis.ticks = element_blank(),
        axis.text.x = element_text(angle = 30, vjust = 0)) 

# FIXME, this takes a long time 
p

ggsave(here::here("figs/counts-per-year.pdf"), p, height=3, width=6)

Requests by state population

library(scales)


dcounts %>% 
  filter(chamber == "Senate") %>% 
    group_by(member_state, year, pop2010) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>% 
    group_by(member_state, pop2010) %>%
  summarise(mean = mean(n)) %>% ungroup() %>% 
    ggplot() + 
    geom_point(aes(x = pop2010, y = mean), color = "light blue") + 
    geom_smooth(aes(x = pop2010, y = mean)) + 
  geom_text_repel(aes(x = pop2010, 
                y = mean, 
                label = ifelse(mean > 260 |# & log(pop2010) < 17.4 | 
                                 mean < 80 & (log(pop2010) > 17 | log(pop2010) <14), 
                               member_state, 
                               "")), 
            seed = 44,
            check_overlap = T, 
            min.segment.length = 0,
            size = 2.5) + theme_bw() +
  labs(title = "",
       x = "State Population (log scale)",
       y = "Average Number of Requests per Year") +
  scale_x_log10(labels = comma)

---

Types of requests

81% of the average legislators’ contacts with agencies are about constituency service.

19% of the average legislators’ contacts with agencies are about policy work.

type	n
Constituency Service	299741
Policy Work	71514
NA	99832

We classify legislator requests into five overall types:

All data:

# paper figure 
type_totals <- all_contacts2 %>% 
  filter(Type != "To be coded") %>%  
  mutate(total = n()) %>% 
  group_by(Type, total, type) %>% 
  summarise(nT = n()) %>% 
  mutate(percent = paste0(nT, " = ", 100*round(nT/total, 2), "%"),
         Type = str_c(Type, "\n    ", percent)) %>% 
  ungroup() %>% 
  group_by(type) %>% 
  mutate(nt = sum(nT),
         percent_t = paste0(100*round(nt/total, 2), "%"))

type_totals %<>% mutate(Type = Type %>% 
                          str_replace_all("\\)", "") %>% 
                          str_replace_all(" \\(", ",\n") )

# horizontal
type_totals %>% 
  ggplot() + 
  geom_col(alpha = .7,aes(x = Type, fill = type, y = nT) ) + 
  scale_y_continuous(breaks = seq(0,500000, 50000), limits = c(0,300000)) + 
  labs(x = "",
       fill = "",
       y = paste("Number of Contacts"),
       title = "") +
  #geom_text(aes(label = Type, x = Type,y = nT) ) + 
  theme(panel.border = element_blank(),
        panel.grid.major.x = element_blank(),
        axis.ticks = element_blank(),
        #axis.text.x.top = element_text(),
        legend.position ="none") + 
  ggpubr::geom_bracket(
    xmin = type_totals$Type[1], xmax = type_totals$Type[3], 
    y.position = 260000,
    label = paste(type_totals$percent_t[1], "Constituency Service" ),
    #hjust = 1,
    vjust = -1
  ) + 
  ggpubr::geom_bracket(
    xmin = type_totals$Type[4], xmax = type_totals$Type[5], 
    y.position = 75000,
    label = paste(type_totals$percent_t[5], "Policy Work"),
    #hjust = 1,
    vjust = -1
  )+ 
  scale_fill_viridis_d(option = "cividis", end = .8, direction = -1)

all_contacts2 %>% 
  filter(Type != "To be coded") %>%  
  mutate(total = n()) %>% 
  group_by(Type, total, type) %>% 
  summarise(nT = n()) %>% 
  mutate(percent = paste0(nT, " = ", 100*round(nT/total, 2), "%"),
         Type = str_c(Type, "\n    ", percent)) %>% 
  ungroup() %>% 
  #distinct(type, Type, nT)
  ggplot() + 
  geom_col(alpha = .7,aes(x = type, fill = Type, y = nT), position ="stack") + 
  #coord_flip() + 
  scale_y_continuous(labels = scales::comma) + 
  labs(x = "",
       fill = "",
       y = paste("Number of Contacts\nN =", sum(nrow(all_contacts2))),
       title = "") +
  #theme_void() + 
  theme(panel.border = element_blank(),
        panel.grid.major.x = element_blank(),
        axis.ticks = element_blank(),
        axis.text.x = element_text())

Horizontal format:

type_totals %<>% 
  #distinct(type, Type, nT)
  mutate(Type = Type %>% str_replace(",\n", ", "))

type_totals %>% 
  ggplot() + 
  geom_col(alpha = .7,aes(x = Type, fill = type, y = nT/1000) ) + 
  coord_flip() + 
  scale_y_continuous(breaks = seq(0,500, 50), limits = c(0,500)) + 
  labs(x = "",
       fill = "",
       y = paste("Number of Contacts (Thousands)"),
       title = "") +
  #theme_void() + 
  theme(panel.border = element_blank(),
        panel.grid.major.y = element_blank(),
        axis.ticks = element_blank(),
        axis.text.x.top = element_text(),
        legend.position ="none") + 
  ggpubr::geom_bracket(
    xmin = type_totals$Type[1], xmax = type_totals$Type[3], y.position = 300,
    label = paste(type_totals$percent_t[1], type_totals$type[1]),
    hjust = -.3,
    vjust = 1.5,
    label.size = 3.5
  ) + 
  ggpubr::geom_bracket(
    xmin = type_totals$Type[4], xmax = type_totals$Type[5], y.position = 100,
    label = paste(type_totals$percent_t[5], type_totals$type[5]),
    hjust = -.3,
    vjust = 1,
    label.size = 3.5
  )

all_contacts2 %>% 
  filter(Type != "To be coded") %>%  
  mutate(total = n()) %>% 
  group_by(Type, total, type) %>% 
  summarise(nT = n()) %>% 
  mutate(percent = paste0(nT, " = ", 100*round(nT/total, 2), "%"),
         Type = str_c(Type, "\n    ", percent)) %>% 
  ungroup() %>% 
  #distinct(type, Type, nT)
  ggplot() + 
  geom_col(alpha = .7,aes(x = type, fill = Type, y = nT/1000), position ="stack") + 
  coord_flip() + 
  scale_y_continuous(breaks = seq(0,500, 50)) + 
  labs(x = "",
       fill = "",
       y = paste("Number of Contacts (Thousands)"),
       title = "") +
  #theme_void() + 
  theme(panel.border = element_blank(),
        panel.grid.major.y = element_blank(),
        axis.ticks = element_blank(),
        axis.text.x.top = element_text())

2007-2017 data:

N <- all_contacts2 %>% 
  filter(year <2018,
         year > 2006) %>% 
  nrow()

all_contacts2 %>% 
  filter(Type != "To be coded",
         year <2018,
         year > 2006) %>%  
    mutate(total = n()) %>% 
  group_by(Type, total, type) %>% 
  summarise(nT = n()) %>% 
  mutate(percent = paste0(nT, " = ", 100*round(nT/total, 2), "%"),
         Type = str_c(Type, "\n    ", percent)) %>% 
  ungroup() %>% 
  #distinct(type, Type, nT)
  ggplot() + 
  geom_col(alpha = .7,aes(x = type, fill = Type, y = nT/1000), position ="stack") + 
  coord_flip() + 
  scale_y_continuous(breaks = seq(0,500, 50)) + 
  labs(x = "",
       fill = "",
       y = paste("Number of Contacts (Thousands)\nN =", N),
       title = "") +
  #theme_void() + 
  theme(panel.border = element_blank(),
        panel.grid.major.y = element_blank(),
        axis.ticks = element_blank(),
        axis.text.x.top = element_text())

---

Requests by institutional position

The first plot is the contacts we have coded. The second assumes these ratios hold for uncoded letters.

chairs <- all_contacts %>% distinct(icpsr, chair, year) 

d %<>% left_join(chairs) %>% 
  mutate(chair = chair %>%
           replace_na(0),
         Chair = ifelse(chair == 1, "Committee Chair", " Non-Chair"))

chair <- d %>% 
  group_by(type, year, icpsr, Chair) %>% 
  summarise(n = sum(per_icpsr_chamber_year_agency_type)) %>%
  group_by(type, Chair) %>% 
  summarise(mean = mean(n)) %>% 
  ungroup() %>%
  filter(!is.na(type))

chair %>% 
  ggplot()  + 
  aes(x = type, y = mean) +
  geom_col(alpha = .7,position = "dodge") + 
  facet_wrap("Chair") + 
  labs(x = "", # "Type of Legislator Request to Agency",
       y = "Average Requests per Year") +
  theme(panel.grid.major.x = element_blank(),
        panel.border = element_blank(),
        axis.ticks.x = element_blank())

Change in Levels of Constituency Service and Policy Work as Legislators Gain Power

chair %>% 
  # assuming consistant proportions to uncoded data 
  mutate(mean = mean*multiplier) %>% 
  ggplot()  + 
  aes(x = type, y = mean) +
  geom_col(alpha = .7,position = "dodge") + 
  facet_wrap("Chair") + 
  labs(x = "Type of Legislator Request to Agency",
       y = "Average Requests per Year") +
  theme(panel.grid.major.x = element_blank(),
        panel.border = element_blank(),
        axis.ticks.x = element_blank())

Change in Levels of Constituency Service and Policy Work as Legislators Gain Power

Level

Member-level

TODO

• [ x ] members.Rdata going back to most senior member’s first Congress in “data/nominate.Rdata”
• [ x ] what to do with party and chamber switchers? Drop them from counts.

#load(here::here('data', 'dcounts_min.Rdata'))

##creating the aggregate count variable
## one obs per agency, member, year 
d <- dcounts_min %>% 
  group_by(agency, icpsr, chamber, year) %>% 
  summarise(perYear = sum(per_icpsr_chamber_year_agency_type)) 

# dcounts_min %>% 
#   mutate(type2 = ifelse(TYPE %in% c(1,2,3), "policy", NA ),
#          type2 = ifelse(TYPE %in% c(4,5), "constituent", type2)) %>% 
#   group_by(type2, agency, icpsr, chamber, year) %>% 
#   # 2-type counts 
#   mutate(perYear_type2 = sum(per_icpsr_chamber_year_agency_type),
#          # versions with NAs for models
#          perYear_con = ifelse(type2 == "constituent", perYear_type2),
#          perYear_pol = ifelse(type2 == "policy", perYear_type2))
#   

# constituency service
d_sub <- dcounts_min %>%
  filter(TYPE %in% c(1, 2, 3)) %>%
  group_by(agency, icpsr, chamber, year) %>%
  summarise(perYear_con = sum(per_icpsr_chamber_year_agency_type))

# policy
d_sub2<- dcounts_min %>%
  subset(TYPE %in% c(4, 5)) %>%
  group_by(agency, icpsr, chamber, year) %>%
  summarise(perYear_pol = sum(per_icpsr_chamber_year_agency_type))


d %<>% left_join(d_sub) %>% left_join(d_sub2)

# congress
d$congress <- year_congress(d$year)

d %<>% left_join(members %>% 
                 select(congress, icpsr, party,party_code, state_abbrev, district_code, nominate.dim1, presidents_party, female, chair, ranking_minority, party_leader, party_whip, majority, prestige, prestige_chair, yearelected, state), by = c('congress', 'icpsr'))

#FIXME this should all be in members data 
# nom <- read_csv(here::here('data', 'nominate_data.csv'))
nom2<- members %>% 
  group_by(icpsr) %>% 
  summarise(first_cong = min(congress))

nom2$first_year <- 1787 + 2*nom2$first_cong

df<- left_join(d, nom2, by ='icpsr' )
#/FIXME 

# Tenure vars 
df$tenure<- df$year - df$first_year
df$first<- ifelse(df$tenure==0, 1, 0)
df$second<- ifelse(df$tenure==1, 1, 0)
df$third<- ifelse(df$tenure==2, 1, 0)
df$fourth<- ifelse(df$tenure==3, 1, 0)
df$fifth<- ifelse(df$tenure==4, 1, 0)
df$sixth<- ifelse(df$tenure==5, 1, 0)

# colnames(agency_vars)
df %<>% 
  #select(-oversight_committee, -oversight_committee_chair) %>% #FIXME these should be the same
  left_join(agency_vars %>% 
                    select(agency, icpsr, chamber, year, oversight_committee,
                           oversight_committee_chair), by = c('agency', 'icpsr', 'year', 'chamber')
            )

df$icpsr_agency<- paste(df$agency, df$icpsr, sep='_')
df$agency_year<- paste(df$agency, df$year, sep='_')

# tenure 
df %<>% ungroup() %>% 
  group_by(icpsr) %>% 
  mutate(max_year = max(tenure)) 

df$survive <- ifelse((df$chamber=='House' & df$max_year>1)| (df$chamber=='Senate' & df$max_year>5),1, 0 )


##purging repeated values 
#FIXME repeats are probably chamber switchers
df %<>% 
  add_count(icpsr, year, agency, chamber, 
            name = "n") 

#FIXME !!! 
doubles <- filter(df, n>1) # no doubles (doubles were chamber-switchers)

doubles %>% kablebox()

agency	icpsr	chamber	year	perYear	perYear_con	perYear_pol	congress	party	party_code	state_abbrev	district_code	nominate.dim1	presidents_party	female	chair	ranking_minority	party_leader	party_whip	majority	prestige	prestige_chair	yearelected	state	first_cong	first_year	tenure	first	second	third	fourth	fifth	sixth	oversight_committee	oversight_committee_chair	icpsr_agency	agency_year	max_year	survive	n
Amtrak	14435	House	2013	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	Amtrak_14435	Amtrak_2013	21	1	2
Amtrak	14435	House	2013	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	Amtrak_14435	Amtrak_2013	21	1	2
Amtrak	14435	House	2014	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	Amtrak_14435	Amtrak_2014	21	1	2
Amtrak	14435	House	2014	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	Amtrak_14435	Amtrak_2014	21	1	2
Amtrak	14435	Senate	2013	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	Amtrak_14435	Amtrak_2013	21	1	2
Amtrak	14435	Senate	2013	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	Amtrak_14435	Amtrak_2013	21	1	2
Amtrak	14435	Senate	2014	1	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	0	0	Amtrak_14435	Amtrak_2014	21	1	2
Amtrak	14435	Senate	2014	1	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	0	0	Amtrak_14435	Amtrak_2014	21	1	2
Amtrak	20115	House	2009	0	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	Amtrak_20115	Amtrak_2009	15	1	2
Amtrak	20115	House	2009	0	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	Amtrak_20115	Amtrak_2009	15	1	2
Amtrak	20115	House	2010	1	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	9	0	0	0	0	0	0	0	0	Amtrak_20115	Amtrak_2010	15	1	2
Amtrak	20115	House	2010	1	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	9	0	0	0	0	0	0	0	0	Amtrak_20115	Amtrak_2010	15	1	2
Amtrak	20115	Senate	2009	0	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	Amtrak_20115	Amtrak_2009	15	1	2
Amtrak	20115	Senate	2009	0	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	Amtrak_20115	Amtrak_2009	15	1	2
Amtrak	20115	Senate	2010	0	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	9	0	0	0	0	0	0	NA	NA	Amtrak_20115	Amtrak_2010	15	1	2
Amtrak	20115	Senate	2010	0	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	9	0	0	0	0	0	0	NA	NA	Amtrak_20115	Amtrak_2010	15	1	2
Amtrak	20730	House	2011	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	Amtrak_20730	Amtrak_2011	11	1	2
Amtrak	20730	House	2011	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	Amtrak_20730	Amtrak_2011	11	1	2
Amtrak	20730	House	2012	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	Amtrak_20730	Amtrak_2012	11	1	2
Amtrak	20730	House	2012	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	Amtrak_20730	Amtrak_2012	11	1	2
Amtrak	20730	Senate	2011	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	Amtrak_20730	Amtrak_2011	11	1	2
Amtrak	20730	Senate	2011	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	Amtrak_20730	Amtrak_2011	11	1	2
Amtrak	20730	Senate	2012	1	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	0	0	Amtrak_20730	Amtrak_2012	11	1	2
Amtrak	20730	Senate	2012	1	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	0	0	Amtrak_20730	Amtrak_2012	11	1	2
Amtrak	20735	House	2009	1	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	0	0	Amtrak_20735	Amtrak_2009	11	1	2
Amtrak	20735	House	2009	1	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	0	0	Amtrak_20735	Amtrak_2009	11	1	2
Amtrak	20735	House	2010	0	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	NA	NA	Amtrak_20735	Amtrak_2010	11	1	2
Amtrak	20735	House	2010	0	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	NA	NA	Amtrak_20735	Amtrak_2010	11	1	2
Amtrak	20735	Senate	2009	0	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	NA	NA	Amtrak_20735	Amtrak_2009	11	1	2
Amtrak	20735	Senate	2009	0	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	NA	NA	Amtrak_20735	Amtrak_2009	11	1	2
Amtrak	20735	Senate	2010	2	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	0	0	Amtrak_20735	Amtrak_2010	11	1	2
Amtrak	20735	Senate	2010	2	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	0	0	Amtrak_20735	Amtrak_2010	11	1	2
Amtrak	29534	House	2007	0	0	0	110		200	MS	0	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	10	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2007	21	1	2
Amtrak	29534	House	2007	0	0	0	110		200	MS	1	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	10	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2007	21	1	2
Amtrak	29534	House	2008	0	0	0	110		200	MS	0	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2008	21	1	2
Amtrak	29534	House	2008	0	0	0	110		200	MS	1	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2008	21	1	2
Amtrak	29534	Senate	2007	0	0	0	110		200	MS	0	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	10	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2007	21	1	2
Amtrak	29534	Senate	2007	0	0	0	110		200	MS	1	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	10	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2007	21	1	2
Amtrak	29534	Senate	2008	0	0	0	110		200	MS	0	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2008	21	1	2
Amtrak	29534	Senate	2008	0	0	0	110		200	MS	1	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	Amtrak_29534	Amtrak_2008	21	1	2
CNCS	14435	House	2013	1	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	0	0	CNCS_14435	CNCS_2013	21	1	2
CNCS	14435	House	2013	1	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	0	0	CNCS_14435	CNCS_2013	21	1	2
CNCS	14435	House	2014	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CNCS_14435	CNCS_2014	21	1	2
CNCS	14435	House	2014	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CNCS_14435	CNCS_2014	21	1	2
CNCS	14435	Senate	2013	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	CNCS_14435	CNCS_2013	21	1	2
CNCS	14435	Senate	2013	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	CNCS_14435	CNCS_2013	21	1	2
CNCS	14435	Senate	2014	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CNCS_14435	CNCS_2014	21	1	2
CNCS	14435	Senate	2014	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CNCS_14435	CNCS_2014	21	1	2
CNCS	20115	House	2009	0	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2009	15	1	2
CNCS	20115	House	2009	0	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2009	15	1	2
CNCS	20115	House	2010	0	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	9	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2010	15	1	2
CNCS	20115	House	2010	0	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	9	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2010	15	1	2
CNCS	20115	Senate	2009	0	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2009	15	1	2
CNCS	20115	Senate	2009	0	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	8	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2009	15	1	2
CNCS	20115	Senate	2010	0	0	0	111		200	IL	0	0.274	0	0	0	0	0	0	NA	1	0	NA	illinois	107	2001	9	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2010	15	1	2
CNCS	20115	Senate	2010	0	0	0	111		200	IL	10	0.274	0	0	0	0	0	0	0	1	0	2008	illinois	107	2001	9	0	0	0	0	0	0	NA	NA	CNCS_20115	CNCS_2010	15	1	2
CNCS	20730	House	2011	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CNCS_20730	CNCS_2011	11	1	2
CNCS	20730	House	2011	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CNCS_20730	CNCS_2011	11	1	2
CNCS	20730	House	2012	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CNCS_20730	CNCS_2012	11	1	2
CNCS	20730	House	2012	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CNCS_20730	CNCS_2012	11	1	2
CNCS	20730	Senate	2011	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CNCS_20730	CNCS_2011	11	1	2
CNCS	20730	Senate	2011	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CNCS_20730	CNCS_2011	11	1	2
CNCS	20730	Senate	2012	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CNCS_20730	CNCS_2012	11	1	2
CNCS	20730	Senate	2012	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CNCS_20730	CNCS_2012	11	1	2
CNCS	20735	House	2009	0	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	NA	NA	CNCS_20735	CNCS_2009	11	1	2
CNCS	20735	House	2009	0	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	NA	NA	CNCS_20735	CNCS_2009	11	1	2
CNCS	20735	House	2010	0	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	NA	NA	CNCS_20735	CNCS_2010	11	1	2
CNCS	20735	House	2010	0	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	NA	NA	CNCS_20735	CNCS_2010	11	1	2
CNCS	20735	Senate	2009	0	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	NA	NA	CNCS_20735	CNCS_2009	11	1	2
CNCS	20735	Senate	2009	0	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	2	0	0	1	0	0	0	NA	NA	CNCS_20735	CNCS_2009	11	1	2
CNCS	20735	Senate	2010	4	0	0	111		100	NY	0	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	0	0	CNCS_20735	CNCS_2010	11	1	2
CNCS	20735	Senate	2010	4	0	0	111		100	NY	20	-0.474	1	1	0	0	0	0	1	1	0	2008	new york	110	2007	3	0	0	0	1	0	0	0	0	CNCS_20735	CNCS_2010	11	1	2
CSOSA	14435	House	2013	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2013	21	1	2
CSOSA	14435	House	2013	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2013	21	1	2
CSOSA	14435	House	2014	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2014	21	1	2
CSOSA	14435	House	2014	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2014	21	1	2
CSOSA	14435	Senate	2013	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2013	21	1	2
CSOSA	14435	Senate	2013	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2013	21	1	2
CSOSA	14435	Senate	2014	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2014	21	1	2
CSOSA	14435	Senate	2014	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	CSOSA_14435	CSOSA_2014	21	1	2
CSOSA	20730	House	2011	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CSOSA_20730	CSOSA_2011	11	1	2
CSOSA	20730	House	2011	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CSOSA_20730	CSOSA_2011	11	1	2
CSOSA	20730	House	2012	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CSOSA_20730	CSOSA_2012	11	1	2
CSOSA	20730	House	2012	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CSOSA_20730	CSOSA_2012	11	1	2
CSOSA	20730	Senate	2011	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CSOSA_20730	CSOSA_2011	11	1	2
CSOSA	20730	Senate	2011	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	4	0	0	0	0	1	0	NA	NA	CSOSA_20730	CSOSA_2011	11	1	2
CSOSA	20730	Senate	2012	0	0	0	112		200	NV	0	0.463	0	0	0	0	0	0	0	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CSOSA_20730	CSOSA_2012	11	1	2
CSOSA	20730	Senate	2012	0	0	0	112		200	NV	2	0.463	0	0	0	0	0	0	1	1	0	2010	nevada	110	2007	5	0	0	0	0	0	1	NA	NA	CSOSA_20730	CSOSA_2012	11	1	2
CSOSA	29534	House	2008	0	0	0	110		200	MS	0	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	CSOSA_29534	CSOSA_2008	21	1	2
CSOSA	29534	House	2008	0	0	0	110		200	MS	1	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	CSOSA_29534	CSOSA_2008	21	1	2
CSOSA	29534	Senate	2008	0	0	0	110		200	MS	0	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	CSOSA_29534	CSOSA_2008	21	1	2
CSOSA	29534	Senate	2008	0	0	0	110		200	MS	1	0.376	1	NA	0	0	0	0	0	1	0	2006	mississippi	105	1997	11	0	0	0	0	0	0	NA	NA	CSOSA_29534	CSOSA_2008	21	1	2
DHHS_ACF	14435	House	2013	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	DHHS_ACF_14435	DHHS_ACF_2013	21	1	2
DHHS_ACF	14435	House	2013	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	DHHS_ACF_14435	DHHS_ACF_2013	21	1	2
DHHS_ACF	14435	House	2014	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	DHHS_ACF_14435	DHHS_ACF_2014	21	1	2
DHHS_ACF	14435	House	2014	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	NA	NA	DHHS_ACF_14435	DHHS_ACF_2014	21	1	2
DHHS_ACF	14435	Senate	2013	0	0	0	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	DHHS_ACF_14435	DHHS_ACF_2013	21	1	2
DHHS_ACF	14435	Senate	2013	0	0	0	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	16	0	0	0	0	0	0	NA	NA	DHHS_ACF_14435	DHHS_ACF_2013	21	1	2
DHHS_ACF	14435	Senate	2014	1	0	1	113		100	MA	0	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	0	0	DHHS_ACF_14435	DHHS_ACF_2014	21	1	2
DHHS_ACF	14435	Senate	2014	1	0	1	113		100	MA	5	-0.511	1	0	0	1	0	0	0	1	0	2012	massachusetts	105	1997	17	0	0	0	0	0	0	0	0	DHHS_ACF_14435	DHHS_ACF_2014	21	1	2

# drop doubles 
df %<>% filter(n<2) 


# LEVEL MODELS

df %<>% 
  # mutate(agency_year) %>% 
  mutate(lpyear = log(perYear + 1))

---

District-level

• Were most maps in place by 2011?

####
# Redistricting? were most maps in place by 2011?
decade<- case_when(df$year < 2011 ~ '0', 
                   df$year > 2010 ~ '1')

state_dist<- case_when(df$chamber=='Senate'~ paste(df$state,
                                                    df$district_code, 
                                                    sep='_' ), 
                       df$chamber =='House'~ paste(paste(df$state, 
                                                          df$district_code, 
                                                          sep='_'), 
                                                    decade,
                                                    sep='_'))




df$state_dist<- state_dist

perDist<- df %>% 
  group_by(year, state_dist, icpsr, chamber, tenure, state) %>% 
  summarise(perYear = sum(perYear)) %>%
  distinct()



## Now, we want to identify when there is a new member in a district.  To do this, we look for changes in the icpsr number
perDist$new_member<- ifelse(perDist$tenure==0, 1, 0)

perDist$second_year<- ifelse(perDist$tenure==1, 1, 0)
perDist$third_year<- ifelse(perDist$tenure==2, 1, 0)
perDist$fourth_year<- ifelse(perDist$tenure==3, 1, 0)
perDist$fifth_year<- ifelse(perDist$tenure==4, 1, 0)
perDist$sixth_year<- ifelse(perDist$tenure==5, 1, 0)


state_level<- perDist %>% group_by(state, year) %>% 
  summarise(numpers = n(), 
            sumYear = sum(perYear), 
            mean_new = mean(new_member)) 

state_new <- ifelse(state_level$mean_new>0, 1, 0)

state_level$state_new <- state_new

perDist %<>% left_join(state_level %>% 
                         select(state_new, mean_new, state, year), 
                    by = c('state', 'year'))
#########################
# saved as DistrictLevel
########################
# save(perDist, file =  here::here("data", "perDist.Rdata"))
############################################

TODO

Priority

##creating the ratio variable 
d_rat <- df %>% 
  group_by(year, icpsr, chamber,
           prestige, prestige_chair, chair, ranking_minority, majority, presidents_party,
           first, second, third, fourth, fifth, sixth) %>% 
  # per member per year counts per type2
  summarise(perCon = sum(perYear_con), 
            perPol = sum(perYear_pol)) %>% 
  # ratio of con to pol 
  #FIXME switch? 
  mutate(ratio = perCon/(perCon + perPol))

# drop doubles
d_rat %<>% group_by(year, icpsr, chamber) %>% add_count(name = "n")

doubles_ratio <- filter(d_rat, n>1)

doubles_ratio %>% kablebox()

year	icpsr	chamber	prestige	prestige_chair	chair	ranking_minority	majority	presidents_party	first	second	third	fourth	fifth	sixth	perCon	perPol	ratio	n

d_rat %<>% filter(n<2) 

############################################
# d_rat saved as ProportionContact.dta')
# save(d_rat, file =  here::here("data", "d_rat.Rdata"))
############################################

# RATIO MODEL 

TODO

Joint Effect of Capacity and Priority

We observe the total requests, \(z\), and ratio, \(r\), of policy work to constituency service. \(z\) is the sum of policy requests \(x\) and constituency service requests \(y\). \(r\) is \(x/y\)./

r = .2

r2 = .31

z = 100

x = z*r
y = z-z*r


z2 = 120

x2 = z2*r2
y2 = z2-z2*r2

Combining model estimates for the shift in the level of requests and relative priority of policy work to constituency service, we see that even as the ratio of policy work increases, constituency service also significantly increases.

These estimates are calculated by adding diff-in-diff regression coefficients for committee chairs’ increased levels of requests to the baseline levels for non-chairs. We also adjust the ratio of policy work to constituency service based on the diff-in-diff regression coefficients for committee chairs’ increased policy work ratio.

n_agencies <- dcounts_min$agency %>% unique() %>% length()

#FIXME replace with tidy model output 
# TABLE 1
level <- tribble(~term, ~estimate, ~std.error,
                 "chair", 0.206, 0.0951)

# per member across all agencies 
level$estimate <- level$estimate*n_agencies
level$std.error <- level$std.error*n_agencies

# ratio of policy work for non-chairs
base_ratio <- chair$mean[3]/chair$mean[1]

# level estimates with baseline ratio (for presenting level effects only)
level_con <- level %>% mutate(estimate = estimate*(1-base_ratio),
                               std.error = std.error*(1-base_ratio))

level_pol <- level %>% mutate(estimate = estimate*(base_ratio),
                               std.error = std.error*(base_ratio))


# Priority Table 
priority <- tribble(~term, ~estimate, ~std.error,
                 "chair", 0.0875, 0.0175)

# adjust level estimates with new ratio (joint effects )
new_ratio <- base_ratio + priority$estimate[1]

level_con <- level %>% mutate(estimate = estimate*(1-new_ratio),
                               std.error = std.error*(1-new_ratio))

level_pol <- level %>% mutate(estimate = estimate*(new_ratio),
                               std.error = std.error*(new_ratio))


#FIXME, include priority model uncertianty too
chair$std.error <- NA
# constituent error 
chair$std.error[2] <- level_con$std.error[1]
# policy error
chair$std.error[4] <- level_pol$std.error[1]


# chair mean = non-chair mean * level coef
# constituent estimate 
chair$mean[2] <- chair$mean[1] + (1 + level_con$estimate[1] )
# policy estimate
chair$mean[4] <- chair$mean[3] + (1+level_pol$estimate[1] )

chair_adjusted <- chair %>% 
  # assuming consistant proportions to uncoded data 
  mutate(mean = mean*multiplier) %>%
  # baseline = non-chairs mean
  mutate(baseline = ifelse(Chair == "Committee Chair", NA, mean) ) %>%
  fill(baseline) %>%
  # point estimates and uncertiatty only for chair estimates 
  mutate(estimate = ifelse(Chair == "Committee Chair", mean, NA)) %>% 
  mutate(std.error = ifelse(Chair == "Committee Chair", std.error, NA)) 

chair_adjusted %>% 
  #mutate(Chair = ifelse() )
  ggplot()  + 
  aes(x = type, y = mean) +
  geom_col(alpha = .7, position = "dodge") + 
  geom_hline(aes(yintercept = baseline), linetype = 2) + 
  geom_pointrange(aes(y = estimate,
                      ymin = mean-1.96*std.error,
                      ymax = mean+1.96*std.error)) + 
  facet_wrap("Chair") + 
  labs(x = "", #Type of Legislator Request to Agency",
       y = "Average Requests per Year") +
  theme(panel.grid.major.x = element_blank(),
        panel.border = element_blank(),
        axis.ticks.x = element_blank())

Difference-in-Difference Estimates of Change in Levels of Constituency Service and Policy Work as Legislators Gain Power

chair_adjusted %<>% mutate(baseline = ifelse(type =="Constituency\nService", baseline+ 12.4, baseline))


# stacked to show ratio with PERCENTS 
chair_adjusted %>% 
  #mutate(Chair = ifelse() )
  ggplot()  + 
  aes(x = Chair, y = mean, fill = type) +
  geom_col(alpha = .7, position = "fill") +  
  facet_wrap("Chair", scales = "free_x") + 
  labs(x = "", #Type of Legislator Request to Agency",
       y = "Share of Requests per Year",
       fill = "") +
  theme(panel.grid.major.x = element_blank(),
        axis.text.x = element_blank(),
        panel.border = element_blank(),
        axis.ticks.x = element_blank())

Difference-in-Difference Estimates of Change in Levels of Constituency Service and Policy Work as Legislators Gain Power

# stacked to show ratio + free scales 
chair_adjusted %>% 
  #mutate(Chair = ifelse() )
  ggplot()  + 
  aes(x = Chair, y = mean, fill = type) +
  geom_col(alpha = .7, position = "stack") + 
  #geom_hline(aes(yintercept = baseline), linetype = 2) + 
  #geom_pointrange(aes(y = estimate,
                      #ymin = mean-1.96*std.error,
                      #ymax = mean+1.96*std.error)) + 
  facet_wrap("Chair", scales = "free") + 
  #scale_y_
  labs(x = "", #Type of Legislator Request to Agency",
       y = "Average Requests per Year",
       fill = "") +
  theme(panel.grid.major.x = element_blank(),
        axis.text.x = element_blank(),
        panel.border = element_blank(),
        axis.ticks.x = element_blank())

Difference-in-Difference Estimates of Change in Levels of Constituency Service and Policy Work as Legislators Gain Power

# stacked to show ratio + CI
chair_adjusted %>% 
  #mutate(Chair = ifelse() )
  ggplot()  + 
  aes(x = Chair, y = mean, fill = type) +
  geom_col(alpha = .7, position = "stack") + 
  geom_hline(aes(yintercept = baseline), linetype = 2) + 
  geom_pointrange(aes(y = estimate,
                      ymin = mean-1.96*std.error,
                      ymax = mean+1.96*std.error)) + 
  facet_wrap("Chair", scales = "free") + 
  labs(x = "", #Type of Legislator Request to Agency",
       y = "Average Requests per Year",
       fill = "") +
  theme(panel.grid.major.x = element_blank(),
        axis.text.x = element_blank(),
        panel.border = element_blank(),
        axis.ticks.x = element_blank())

Difference-in-Difference Estimates of Change in Levels of Constituency Service and Policy Work as Legislators Gain Power

The baseline level (non-chairs) is 55.3494555 constituency service requests and 12.404877 policy requests per year.

The baseline ratio (for non-chairs) of policy work to constituency service is 0.2241192. Adding the model coefficient from the difference in difference model of shifting priorities yields a new ratio of 0.3116192.

The model-estimated level for committee chairs is 72.8128495 constituency service requests and 21.0047737 policy requests per year.

#TODO

• intercoder agrement for USTR

• y-axis label and remove the title from inequality figure

• facet distribution across types by policy and constituent, probably coord flip as well

• [ x ] remove the title from state pop figure